This R21/R33 application responds to NCI program announcement PA-04-095 entitled, "Novel Technologies for In Vivo Imaging" directed toward the development and delivery of innovative, high-risk/high-gain image acquisition and enhancement methods for use in cancer patients. Our grant is designed to achieve specific aims for developing a cardiovascular magnetic resonance (CMR) imaging method for early detection of doxorubicin induced cardiotoxicity. Although doxorubicin prolongs survival and reduces tumor burden for many children and adults with cancer;its use can cause irreversible congestive heart failure (CHF) and death. The most accurate method for detecting doxorubicin cardiotoxicity is an intramyocardial biopsy, but this procedure is invasive, relatively expensive, and not well suited for repetitive examinations. More commonly, serial radionuclide ventriculograms are used to assess left ventricular ejection fraction (LVEF) during the course of doxorubicin therapy. A fall in LVEF indicates those at increased risk for cardiotoxicity, and it is common practice to stop therapy when this occurs. Importantly however, a fall in LVEF may occur only after irreversible cardiac damage has been sustained. For this reason, the total dose of doxorubicin is often limited in cancer patients. A safe, widely available, noninvasive method to detect doxorubicin induced cardiotoxicity before a fall in LVEF occurs would be useful in 2 respects. First, early detection of cardiotoxicity could help prevent the development of CHF, and second, the absence of cardiotoxicity would allow the continuation of therapy so that patients could realize the full benefit of the medication. Although not requested, we provide pilot data indicating that CMR images identify early myocardial injury before the heart sustains a drop in LVEF. In the R21 phase of the award, we describe the software, statistical analyses, and recruitment strategy needed to develop and further test our CMR methodology. In the 3-year R33 phase of the award, we provide the study design to assess the clinical utility of this newly developed CMR methodology. We define the performance targets for determining suitability for transition from the R21 to the R33 phase in a Milestones document that resides between the R21 and R33 application. Each phase (R21 &R33) contains separate sections A-D, and as directed by the instructions, we designate (but do not repeat or re-type) the components of sections A through D of the R21 phase that are applicable to the R33 phase.